Kofax Power PDF OCR: What Tungsten's Editor Does
By Schema LawPublished Updated
The OCR in Kofax Power PDF, now sold as Tungsten Power PDF, reads image-only pages and does two things with the result. Make PDF Searchable adds a hidden text layer behind the scanned page. Convert Assistant turns the pages into editable Word, Excel, PowerPoint or RTF files. Both rely on the same built-in recognition inside the desktop program.
This post describes those features from Tungsten Automation's own help files and product pages, read on October 8, 2026 and linked where they are used. The help cited is for version 2025.3. If the copy on your computer still says Kofax or Nuance on the title bar, menus may sit in slightly different places, and the last section shows where to find the help for your version.
What is Kofax Power PDF called now, and who makes it?
It is called Tungsten Power PDF, and Tungsten Automation makes it. The program reached that name in two steps that the company's own pages record: it was Nuance Power PDF, then Kofax Power PDF, and it took the Tungsten name after Kofax renamed itself in January 2024.
The evidence for each name is on a vendor page:
- Tungsten. The company's announcement, Kofax is now Tungsten Automation, is dated January 16, 2024.
- Kofax. The Power PDF documentation portal lists the current release as Tungsten Power PDF 2025.3 and the previous versions as Kofax Power PDF 5.1.1, 5.1, 5.0, 4.1 and 4.0.
- Nuance. The trial questions on the Power PDF Standard product page mention Nuance Power PDF 2.0 as an older version of the same product.
So "Nuance Power PDF OCR", "Nuance PDF OCR" and "Kofax Power PDF OCR" are searches for one feature in different years of one program. A firm that bought licences once and never upgraded may be running any of the three.
What the Kofax OCR engine is
Tungsten's help does not give the recognition engine a brand name. Its topic on the built-in OCR defines OCR as the process of extracting text from an image and says no more about what is underneath. The help topic on when to use OmniPage says the built-in function is similar to the capabilities available with OmniPage, a separate Tungsten product, and that OmniPage adds manual zoning of pages and more formatting levels for people who want more control.
How does Power PDF make a scanned PDF searchable?
Power PDF runs OCR on the image-only pages and stores the recognized text in a separate layer behind the page image. You start it with Make PDF Searchable on the Home tab, choose which pages to process and a language, and click OK. The page looks the same afterwards.
Tungsten's help topic on making PDF files searchable describes a searchable PDF as one that "presents page images, but also contains the recognized text in a separate layer", and says the format is especially useful for documents that must be archived with their original appearance. An exhibit is that kind of document.
The steps in the open document
- On the Home tab, click Make PDF Searchable.
- In the Convert Pages dialog box, choose whether OCR runs only on pages with image-only parts or on all pages.
- Click Settings and pick the language of the document in the OCR Language list.
- Optionally select Automatically proofread results after OCR.
- Click OK.
Step 2 deserves a pause. The help says that if you choose all pages, "any text layer content previously in the PDF is replaced by the OCR results". On a file that mixes typed pages with scanned ones, the default of image-only pages leaves the typed pages alone. Choose all pages only when the existing text layer is bad and you want it gone.
Settings that change the result
The topic on Searchable PDF Conversion Settings lists these:
| Setting | What the help says it does |
|---|---|
| OCR Language | Tells recognition which language the source is in |
| Reject character | Stands in for a character that could not be recognized. A tilde by default, so REJECT with an unread J appears as RE~ECT |
| Keep original images | Retains the original image after conversion |
| Auto orient pages | Sets the page orientation automatically |
| Process documents using OCR | Converts every page to a bitmap and recognizes it, dropping hidden or covered objects and text in non-standard fonts |
| Look for image-only pages automatically | Checks each PDF as it opens and offers to convert it. Off by default because the check takes time |
The reject character is worth knowing about before you search a converted file. A name with one unread letter is stored with a tilde in it, and a search for the correct spelling will not find it. Searching the finished file for the tilde is a quick way to locate the places recognition gave up.
The proofreader
Power PDF has an interactive check for recognized text. According to the help topic on proofreading and searchable PDF, recognition assigns a confidence level to each character and word and offers the doubted words for review. A panel shows a picture of the word from the page, the current reading, and alternatives from a dictionary. You accept, change or ignore each one. Start it from Home, then Make PDF Searchable, then Proofreader. Whether or not you use it, read names, dates and amounts against the page image before you rely on the text. For what confidence figures mean in general, see the post on OCR confidence scores.
Languages
The help topic on language selection says the program handles over one hundred languages and advises selecting only the languages your document contains. You can select several, with one stated limit: an Asian language cannot be combined with others in one selection. There is an automatic detection option that assigns one language to each full page.
Power PDF Standard versus Advanced for OCR work
Tungsten's help does not mark Make PDF Searchable as an Advanced-only feature. What it does reserve for Power PDF Advanced are the tools around recognition: the Batch Converter, watched folders, redaction, Bates numbering and document management system connections. For one scan at a time, check Standard. For volume and productions, the documentation points to Advanced.
That is a careful sentence because the documentation is written that way. The help has a section called Advanced features in Power PDF that lists what Advanced and Business add. The searchable PDF and Convert Assistant topics sit outside that section. This post does not claim more about Kofax Power PDF Standard OCR than that. If you are buying Standard for recognition, confirm it in the 15-day trial the product page offers.
| Feature | Where Tungsten's help places it |
|---|---|
| Make PDF Searchable in an open document | General help, not in the Advanced section |
| Convert Assistant to Word, Excel, PowerPoint, RTF | General help, not in the Advanced section |
| Batch Converter between TIFF and PDF | Advanced |
| Watched Folder and Scan Inbox | Advanced only |
| Redaction | Advanced |
| Bates numbering | Advanced section |
| Opening from and saving to a document management system | Advanced |
| AI editing with Copilot | Business only |
The product pages agree on the outline. The Power PDF Standard product page describes Advanced as adding batch processing, Bates numbering and stamping, and advanced redaction to everything in Standard, and it lists a one-time price for each edition. This post leaves the figures to that page, because they change. The help says Power PDF Business offers corporate features over the feature set of Advanced. The companion post on choosing a Kofax Power PDF alternative covers the licence terms and what leaving the program would cost you.
Power PDF among other tools
Power PDF is a desktop editor in the same family as the product described in the post on ABBYY FineReader PDF: recognition is one feature among editing, security and conversion tools, and the work happens in a program installed on a computer. The comparison of online OCR and desktop OCR for law firms sets out when that arrangement suits a firm and when a browser service does. Schema Law is the second kind. It converts uploaded scans to text and tables and exports them in the formats on its formats list, and it edits nothing.
Does Power PDF convert a scan to Word or Excel?
Yes. Convert Assistant turns PDF pages, including image-only ones, into Word, Excel, PowerPoint or RTF files, and it recognizes scanned pages with the built-in OCR on the way. You choose a processing mode first, and the mode decides which outputs are offered.
The help topic on Convert Assistant lists four modes:
| Mode | Output | Meant for |
|---|---|---|
| Standard | Word, PowerPoint or RTF | Contracts, letters, articles |
| Legal | Word or RTF | Pleadings and similar court papers |
| Spreadsheet | Excel | Pages that are mainly tables |
| Form | Word or RTF | Static forms you want as fillable Word forms |
Legal mode
Legal mode is the unusual one. The help topic on selecting the processing mode says it handles the furniture of a pleading in specific ways:
- Line numbers in the left margin are kept or dropped, depending on a Retain option.
- The vertical rules beside the line numbers are detected and dropped where possible.
- A header line on each page identifying the case is detected and dropped where possible.
- The caption block naming the parties at the top of the first page is retained with formatting.
- Stamps and signs are usually handled as images, and you choose whether to keep them.
The help adds a practical tip: to reformat a legal document completely, drop the original pleading numbers and recreate them later.
After the searchable pass finishes, search for a margin line number on the pleading, a defined term in the contract, and the exhibit sticker. If the same scan produced a spreadsheet, find an invoice amount there and compare it with the page. Read any broken name on the image before the conversion goes to anyone else.
Image-only pages in a conversion
According to the help topic on image-only PDF files, Convert Assistant shows an icon on each previewed page that tells you whether it has a text layer or is image-only. Under Processing options you decide what happens to the image-only ones. The topic on the Standard panel gives three choices: convert the page with OCR, skip it, or add it to the document as an image. If you skip, the converted file has fewer pages than the source, and the program tells you so at the end. Check that setting before converting a file in which a few scanned pages matter.
Tables to Excel
In Spreadsheet mode you choose how sheets are created: one per table, one per page, or one per document. The help for the Spreadsheet panel says that with one sheet per table, everything that was not detected as a table goes on a separate worksheet named Overview, with links to the table sheets. For a single table on a crowded page, the help describes an Export Area tool under Advanced Processing, then Export, that sends just the area you draw to Excel.
Whatever the mode, the output is a working copy. The post on whether OCR keeps formatting explains which parts of a page tend to survive conversion in any product.
Batch conversion of scanned files in Power PDF
Power PDF offers three ways to process many scans. The Batch Converter turns a folder of TIFF or image-only PDF files into searchable PDFs. A Watched Folder converts files as they arrive. Create Assistant builds searchable PDFs from a list of files. Tungsten's help marks the first two as Power PDF Advanced features.
The Batch Converter
From the help topic on setting up and running batch file conversions:
- Move the files to an input folder.
- On the Advanced Processing tab, select Batch Converter.
- Click Source, choose the folder, and pick PDF or TIFF as the file type.
- Select the files, or Select All.
- Click Destination and choose a different output folder.
- Under Destination, set the file type to Searchable PDF and use Settings for the OCR language.
- Choose a PDF standard in the Compatible with list if you need one, such as PDF/A.
- Run it, then click View Log for a plain text report.
Step 5 says a different folder for a reason. The help warns that if the source and destination folders are the same, "the original files may be overwritten without warning, even if warnings were requested". For anything you received from another party, write the output somewhere else and leave the originals alone.
Watched Folder and Scan Inbox
The help topic on Watched Folders and Scan Inboxes says both are available only in Power PDF Advanced, and that only one of each can be set up. A watched folder monitors a location, converts what lands there and delivers the result to a destination folder. The topic on setting up a watched folder lists Searchable PDF among the output types. A Scan Inbox is different: it notifies you when your scanner drops a file into its target folder, and the help states that no automatic conversion is involved.
The help describes the arrangement: "Typically the watched folder would be at a network location and the destination folder on your local computer." A copier's scan-to-folder share fits that description, with the output type set to Searchable PDF.
Create Assistant
Create Assistant has a profile named Searchable PDF. You select it, add one or more image files or image-only PDFs, set the OCR language, and start PDF creation. The help topic on making PDF files searchable adds that OCR runs only when no accessible text layer is detected in an input file. Where a text layer is found, the existing text is used.
For runs measured in boxes instead of folders, the planning questions are the same in any product. The post on batch OCR for discovery documents covers them.
Where Tungsten publishes the Power PDF help and release notes
Tungsten publishes Power PDF documentation on its DocShield site, one page per version. The page for a version links the release notes, the technical specifications, a getting started guide and the online help. Inside the program, pressing F1 on an open dialog box opens the help topic for that dialog.
What the documentation page for Power PDF 2025.3 offers:
- Release notes for the version.
- Technical specifications, where the help says the list of supported document management systems is kept.
- Getting Started Guide, as a PDF.
- Online help, the source for most of this post, plus separate help for the cloud editor.
- Guides for the Business licence, including a network installation guide.
- Previous versions, a menu that leads to the Kofax-branded documentation for 4.0 through 5.1.1.
Two habits save time. Match the help to the version you run, because a menu path from 2025.3 may not exist in 5.0. And read the release notes before an upgrade, since the vendor's release highlights for Power PDF describe changes to licensing as well as features.